From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Zijlstra Subject: Re: [PATCH v9 3/5] locking/qspinlock: Introduce CNA into the slow path of qspinlock Date: Thu, 23 Jan 2020 11:06:35 +0100 Message-ID: <20200123100635.GE14946@hirez.programming.kicks-ass.net> References: <20200115035920.54451-1-alex.kogan@oracle.com> <20200115035920.54451-4-alex.kogan@oracle.com> <20200123092658.GC14879@hirez.programming.kicks-ass.net> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <20200123092658.GC14879@hirez.programming.kicks-ass.net> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+linux-arm-kernel=m.gmane-mx.org@lists.infradead.org To: Alex Kogan Cc: linux-arch@vger.kernel.org, guohanjun@huawei.com, arnd@arndb.de, dave.dice@oracle.com, jglauber@marvell.com, x86@kernel.org, will.deacon@arm.com, linux@armlinux.org.uk, steven.sistare@oracle.com, linux-kernel@vger.kernel.org, mingo@redhat.com, bp@alien8.de, hpa@zytor.com, longman@redhat.com, tglx@linutronix.de, daniel.m.jordan@oracle.com, linux-arm-kernel@lists.infradead.org List-Id: linux-arch.vger.kernel.org On Thu, Jan 23, 2020 at 10:26:58AM +0100, Peter Zijlstra wrote: > On Tue, Jan 14, 2020 at 10:59:18PM -0500, Alex Kogan wrote: > > +/* this function is called only when the primary queue is empty */ > > +static inline bool cna_try_change_tail(struct qspinlock *lock, u32 val, > > + struct mcs_spinlock *node) > > +{ > > + struct mcs_spinlock *head_2nd, *tail_2nd; > > + u32 new; > > + > > + /* If the secondary queue is empty, do what MCS does. */ > > + if (node->locked <= 1) > > + return __try_clear_tail(lock, val, node); > > + > > + /* > > + * Try to update the tail value to the last node in the secondary queue. > > + * If successful, pass the lock to the first thread in the secondary > > + * queue. Doing those two actions effectively moves all nodes from the > > + * secondary queue into the main one. > > + */ > > + tail_2nd = decode_tail(node->locked); > > + head_2nd = tail_2nd->next; > > + new = ((struct cna_node *)tail_2nd)->encoded_tail + _Q_LOCKED_VAL; > > + > > + if (atomic_try_cmpxchg_relaxed(&lock->val, &val, new)) { > > + /* > > + * Try to reset @next in tail_2nd to NULL, but no need to check > > + * the result - if failed, a new successor has updated it. > > + */ > > I think you actually have an ordering bug here; the load of head_2nd > *must* happen before the atomic_try_cmpxchg(), otherwise it might > observe the new next and clear a valid next pointer. > > What would be the best fix for that; I'm thinking: > > head_2nd = smp_load_acquire(&tail_2nd->next); > > Will? Hmm, given we've not passed the lock around yet; why wouldn't something like this work: smp_store_release(&tail_2nd->next, NULL); if (!atomic_try_cmpxchg_relaxed(&lock, &val, new)) { tail_2nd->next = head_2nd; return false; } The whole second queue is only ever modified by the lock owner, and that is us, so we can pre-terminate the secondary queue (break the circular link), try the cmpxchg and fix it back up when it fails. > > + cmpxchg_relaxed(&tail_2nd->next, head_2nd, NULL); > > + arch_mcs_pass_lock(&head_2nd->locked, 1); > > + return true; > > + } > > + > > + return false; > > +} From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from merlin.infradead.org ([205.233.59.134]:60824 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726026AbgAWKHE (ORCPT ); Thu, 23 Jan 2020 05:07:04 -0500 Date: Thu, 23 Jan 2020 11:06:35 +0100 From: Peter Zijlstra Subject: Re: [PATCH v9 3/5] locking/qspinlock: Introduce CNA into the slow path of qspinlock Message-ID: <20200123100635.GE14946@hirez.programming.kicks-ass.net> References: <20200115035920.54451-1-alex.kogan@oracle.com> <20200115035920.54451-4-alex.kogan@oracle.com> <20200123092658.GC14879@hirez.programming.kicks-ass.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20200123092658.GC14879@hirez.programming.kicks-ass.net> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Alex Kogan Cc: linux@armlinux.org.uk, mingo@redhat.com, will.deacon@arm.com, arnd@arndb.de, longman@redhat.com, linux-arch@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, tglx@linutronix.de, bp@alien8.de, hpa@zytor.com, x86@kernel.org, guohanjun@huawei.com, jglauber@marvell.com, steven.sistare@oracle.com, daniel.m.jordan@oracle.com, dave.dice@oracle.com Message-ID: <20200123100635.mT2PgmqZwQAFSBKWSUMU1Ght3g2YwUOzXhXbVyvbOBA@z> On Thu, Jan 23, 2020 at 10:26:58AM +0100, Peter Zijlstra wrote: > On Tue, Jan 14, 2020 at 10:59:18PM -0500, Alex Kogan wrote: > > +/* this function is called only when the primary queue is empty */ > > +static inline bool cna_try_change_tail(struct qspinlock *lock, u32 val, > > + struct mcs_spinlock *node) > > +{ > > + struct mcs_spinlock *head_2nd, *tail_2nd; > > + u32 new; > > + > > + /* If the secondary queue is empty, do what MCS does. */ > > + if (node->locked <= 1) > > + return __try_clear_tail(lock, val, node); > > + > > + /* > > + * Try to update the tail value to the last node in the secondary queue. > > + * If successful, pass the lock to the first thread in the secondary > > + * queue. Doing those two actions effectively moves all nodes from the > > + * secondary queue into the main one. > > + */ > > + tail_2nd = decode_tail(node->locked); > > + head_2nd = tail_2nd->next; > > + new = ((struct cna_node *)tail_2nd)->encoded_tail + _Q_LOCKED_VAL; > > + > > + if (atomic_try_cmpxchg_relaxed(&lock->val, &val, new)) { > > + /* > > + * Try to reset @next in tail_2nd to NULL, but no need to check > > + * the result - if failed, a new successor has updated it. > > + */ > > I think you actually have an ordering bug here; the load of head_2nd > *must* happen before the atomic_try_cmpxchg(), otherwise it might > observe the new next and clear a valid next pointer. > > What would be the best fix for that; I'm thinking: > > head_2nd = smp_load_acquire(&tail_2nd->next); > > Will? Hmm, given we've not passed the lock around yet; why wouldn't something like this work: smp_store_release(&tail_2nd->next, NULL); if (!atomic_try_cmpxchg_relaxed(&lock, &val, new)) { tail_2nd->next = head_2nd; return false; } The whole second queue is only ever modified by the lock owner, and that is us, so we can pre-terminate the secondary queue (break the circular link), try the cmpxchg and fix it back up when it fails. > > + cmpxchg_relaxed(&tail_2nd->next, head_2nd, NULL); > > + arch_mcs_pass_lock(&head_2nd->locked, 1); > > + return true; > > + } > > + > > + return false; > > +}